ML model for predicting the intensity and frequency of urban building fires

Examining how urbanization affects carbon emissions in the construction industry is crucial for China's efforts to achieve environmentally friendly and low-carbon growth. Nevertheless, current research predominantly employs a unidimensional approach to gauge urbanization, overlooking the nuanced characteristics of novel urbanization. This study explored the spatial and temporal characteristics of the impact of urbanization on carbon emissions from urban residential buildings from a multidimensional perspective, starting from five dimensions: population, economy, land, life, and ecology, using a spatial–temporal geographically weighted regression model (GTWR) and panel data from 30 provinces in China for the period 2010–2020. The findings of this study revealed that, on the whole, population urbanization, land urbanization, and lifestyle urbanization exert a significant facilitating influence on carbon emissions from urban residential buildings. Conversely, economic urbanization exhibits a notable inhibitory effect on carbon emissions from urban residential buildings. Furthermore, ecological urbanization transitions from an inhibitory effect to a promoting effect on carbon emissions from urban residential buildings. The impact of urbanization across various dimensions on carbon emissions from urban residential buildings displays spatiotemporal heterogeneity across different regions and periods. In addition, the structure of energy consumption is a contributing factor to the increase in carbon emissions from urban residential buildings, while there is a notable U-shaped relationship between the level of foreign investment, the industrial structure, and carbon emissions originating from urban residential areas. This study offers profound insights into the relationship between urbanization and carbon emissions, providing valuable guidance for policymakers.

In order to overcome the problems of low energy-saving rate and high net energy consumption existing in the urban residential zero-energy building feasibility model, this paper proposes a new solar source urban residential zero-energy building feasibility model based on Laplace transform method. The Laplacian transform of the energy-saving design parameters of the solar energy source urban residential zero-energy building is used to build a feasibility model of the urban residential zero-energy building. The feasibility parameters of the urban residential zero-energy building are calculated based on the feasibility model, and consumption of building feasibility. The experimental results show that the feasibility model of urban residential zero-energy building construction greatly improves the energy saving rate, the highest energy saving rate can reach 93%, the net energy consumption is reduced, and the minimum energy consumption is only 80 kg/t.

In order to overcome the problems of low energy-saving rate and high net energy consumption existing in the urban residential zero-energy building feasibility model, this paper proposes a new solar source urban residential zero-energy building feasibility model based on Laplace transform method. The Laplacian transform of the energy-saving design parameters of the solar energy source urban residential zero-energy building is used to build a feasibility model of the urban residential zero-energy building. The feasibility parameters of the urban residential zero-energy building are calculated based on the feasibility model, and consumption of building feasibility. The experimental results show that the feasibility model of urban residential zero-energy building construction greatly improves the energy saving rate, the highest energy saving rate can reach 93%, the net energy consumption is reduced, and the minimum energy consumption is only 80 kg/t.

A study on the energy rebound effect of China's residential building energy efficiency

Due to the randomness of interior combustibles, wall thermal inertia, and opening factor, the fire temperature rise in high-rise residential buildings is uncertain. This study investigated 38 urban high-rise residential buildings, created the probability density functions of fire load density, opening factor, and wall thermal inertia, and constructed random fire scenarios for urban high-rise residential buildings. On this basis, relying on the Latin Hypercube Sampling method, this study further explored the probabilistic model for fire temperature rise in urban high-rise residential buildings under the action of uncertain factors, generated the possible temperature rise curves of fires in urban high-rise residential buildings and their probability distribution, and established the most representative temperature rise curve.

Nonlinear influence of urbanization on China's urban residential building carbon emissions: New evidence from panel threshold model

Quantification of the carbon emission of urban residential buildings: The case of the Greater Bay Area cities in China

Comparison of machine learning and logistic regression models in predicting acute kidney injury: A systematic review and meta-analysis

Fire incidents pose a significant global threat, causing extensive damage to human settlements, ecosystems, and economies. The increasing frequency and intensity of fires, exacerbated by climate change, urbanization, and inadequate fire management strategies, have led to rising financial and environmental costs. This literature review examines global fire trends, economic losses, and policy responses, with a focus on how various nations address fire risks through prevention, mitigation, and governance. Developed countries such as the United States, Australia, and members of the European Union have implemented advanced fire suppression technologies, early warning systems, and cross-border cooperation to reduce fire-related risks. In contrast, developing nations, including the Philippines, face challenges in fire prevention due to limited resources, weak enforcement of fire safety regulations, and rapid urban expansion. This literature review also explores the role of the Department of Environment and Natural Resources (DENR) in mitigating fire-related damage in the Philippines through policies on forest conservation, waste management, and climate adaptation. Comparative insights reveal the importance of integrating fire prevention strategies with environmental governance and disaster resilience planning. By analyzing international best practices, this review highlights the need for stronger policy enforcement, public awareness, and adaptive strategies to minimize the economic and ecological impact of fire incidents worldwide.

We report a workflow and the output of a natural language processing (NLP)-based procedure to mine the extant metal–organic framework (MOF) literature describing structurally characterized MOFs and their solvent removal and thermal stabilities. We obtain over 2,000 solvent removal stability measures from text mining and 3,000 thermal decomposition temperatures from thermogravimetric analysis data. We assess the validity of our NLP methods and the accuracy of our extracted data by comparing to a hand-labeled subset. Machine learning (ML, i.e. artificial neural network) models trained on this data using graph- and pore-geometry-based representations enable prediction of stability on new MOFs with quantified uncertainty. Our web interface, MOFSimplify, provides users access to our curated data and enables them to harness that data for predictions on new MOFs. MOFSimplify also encourages community feedback on existing data and on ML model predictions for community-based active learning for improved MOF stability models.

In the context of rapid urbanization in China, urban residential buildings play an important role in the fulfillment of well-being society. Behind the gigantic demand on housings in the future decade, urban residential buildings will exert great impacts on the sustainable development of entire nation which deserves research and analysis. Based on the dichotomized energy types (emergy and operation energy), this study firstly adopts the China economic input-output benchmark of 2007 to estimate the emergy of new-built urban residential buildings. Then, with the end-use energy intensity of urban residences' daily life, the operation energy is figured out and follows energy quantification for building end-of-use phase. Through the complete life-cycle energy consumptions of urban residential buildings, it can be seen that operation energy accounts for 76.5% of total energy use; the energy consumption of construction and end-of-life phases are close whose proportion is approximate 1%; life-cycle energy depends heavily on coal and electricity.

Smart Irrigation System is a complex concept used to control, monitor and automate the irrigation of yields by integrating artificial intelligence techniques such as Machine Learning strategies. SIS has endorsed various machine learning models. However, there has been no attempt to analyze the empirical evidence on ML models in a systematic way. Moreover, ML based SIS often face problems and raise questions that must be resolved. This article presents a systematic literature review of ML based SIS; an overview of the literature on ML is designed, revealing a premier and unbiased survey of the existing empirical research. 55 selected studies published from 2017 to 2023 and nine broadly used ML models were identified. Furthermore, four analysis aspects namely type of ML technique, estimation accuracy, model comparison, and estimation context have been outlined. The findings of this review prove the performance capability of the ML techniques endorsed within SIS. The ML techniques outperform other conventional approaches. However, the application of ML models in SIS is still limited and more effort is needed to obtain well-formed and generalizable results. To this end, and based on the outcomes obtained in this work, future guidelines have been provided to practitioners and researchers to grasp the major contributions and challenges in the state-of-the-art research.

Because of recent advancements in the field of natural language processing (NLP) and machine learning, there is potential to ingest decades of field history and heterogeneous production records. This paper proposes an analytics workflow that leverages artificial intelligence to process thousands of historical workover reports (handwritten and electronic), extract important information, learn patterns in production activity, and train machines to quantify workover impact and derive best practices for field operations. Natural language processing libraries were developed to ingest and catalog gigabytes of field data, identify rich sources of workover information, and extract workover and cost information from unstructured reports. A machine learning (ML) model was developed and trained to predict well intervention categories based on free text describing workovers found in reports. This ML model learnt pattern and context of repeating words pertaining to a workover type (e.g. Artificial Lift, Well Integrity, etc.) and to classify reports accordingly. Statistical models were built to determine return on investment from workovers and rank them based on production improvement and payout time. Today, 80% of an oilfield expert's time can be spent manually organizing data. When processing decades of historical oilfield production data spread across both structured (production timeseries) and unstructured records (e.g., workover reports), experts often face two major challenges: 1) How to rapidly analyze field data with thousands of historical records. 2) How to use the rich historical information to generate effective insights to optimize production. In this paper, we analyzed multiple field datasets in a heterogeneous file environment with 20 different file formats (PDF, Excel, and other formats), 2,000+ files and production history spanning 50+ years across and 2000+ producing wells. Libraries were developed to extract workover files from complex folder hierarchies through an intelligent automated search. Information from reports was extracted through Python libraries and optical character recognition technology to build master data source with production history, workover, and cost information. A neural network model was trained to predict workover class for each report with >85% accuracy. The rich dataset was then used to analyze episodic workover activity by well and compute key performance indicators (KPIs) to identify well candidates for production enhancement. The building blocks included quantifying production upside and calculating return of investment for various workover classes. O&G companies have vast volumes of unstructured data and use less than 1% of it to uncover meaningful insights about field operations. Our workflow describes methodology to ingest both structured and unstructured documents, capture knowledge, quantify production upside, understand capital spending, and learn best practices in workover operations through an automated process. This process helps optimize forward operating expense (OPEX) plan with focus on cost reduction and shortens turnaround time for decision making.

Study question What is the performance of a machine-learning) ML( model in predicting the number of oocytes retrieved during ovarian stimulation, compared to prediction by fertility specialists? Summary answer The ML model consistently outperformed fertility specialists in predicting the number of oocytes in real-world IVF patients treated with r-hFSH-α originator. What is known already Fertility specialists often encounter challenges in accurately predicting oocyte retrieval outcomes before commencing the ovarian stimulation cycle, with frequent under- or overestimations. This variability stems from the subjective interpretation of patient-specific factors, such as age and ovarian reserve markers, during clinical decision-making. While existing research highlights the potential of AI tools to enhance clinical practice, further studies are needed to systematically compare the predictive performance of AI models with the expertise of fertility specialists. Study design, size, duration Twelve real-world IVF patient cases were provided in a web-based survey to 29 experienced fertility specialists from 29 clinics worldwide (Oct–Dec 2024). Patient profile data (age, height, weight, indication for ART treatment, AMH, AFC, freeze-all/fresh transfer) were provided. The r-hFSH-α starting dose was shared with the specialists (assuming no dose adjustment), who were requested to estimate the number of oocytes retrieved. Specialists completed a tutorial and quiz before the first case evaluation. Participants/materials, setting, methods A previously developed XGBoost (advanced tree-based that excels in speed and performance, particularly for structured data) machine learning model prediction of the number of oocytes retrieved according to patient profile and r-hFSH-α starting dose was compared to fertility specialists’ predictions. The predictions were evaluated against the ground truth number of oocytes retrieved. The accuracy of the machine learning model was then compared to specialist predictions. Main results and the role of chance Performance was assessed using mean absolute error (MAE) and mean error (ME) to compare the predictions of the model and fertility specialists against the ground truth number of retrieved oocytes for each case. Physicians achieved an MAE of 3.77 oocytes, compared to an MAE of 1.94 oocytes achieved by the ML model, highlighting the model’s greater accuracy. Physicians achieved an ME of 1.97±4.43 (mean±standard deviation), showing a consistent tendency to underestimate the number of retrieved oocytes. In contrast, the ML model displayed a far more balanced ME of 0.16±2.64, demonstrating significantly lower bias and variability in predictions. To further compare performance, we ranked the physicians and the ML model for each case based on the accuracy of their predictions. A rank of 1 was assigned to the most accurate prediction (closest estimation to the ground-truth per case), while a rank of 30 was assigned to the least accurate. Across all cases, the ML model achieved the best average rank of 8.08, outperforming even the most accurate physicians. The three top-performing physicians had average ranks of 8.17, 10.79, and 10.96. The median physician rank was 15.62, highlighting that the ML model consistently delivered more accurate predictions than the majority of physicians. Limitations, reasons for caution This survey involved a small cohort of reproductive specialists, assessing only 12 clinical cases. Additionally, the machine learning model’s predictions were based exclusively on r-hFSH-α originator-treated cycles. Wider implications of the findings AI models may enhance prediction accuracy and consistency, enabling personalized ovarian stimulation protocols and supporting informed decision-making during patient consultation. Such tools can also provide AI-based second opinion in challenging cases and potentially improve consistency in r-hFSH-α dosing strategy among reproductive specialists. Trial registration number No

Impacts of forest fire frequency on structure and composition of tropical moist deciduous forest communities of Bandhavgarh Tiger Reserve, Central India